The focus of the new work is the Acanthodactylus erythrurus complex, a group built around the red-fringed or fringe-toed lizard formally described back in 1838. Acanthodactylus as a whole is the most species-rich genus in the lizard family Lacertidae, with 44 recognized species, and the erythrurus complex is arguably its most taxonomically tangled branch. The species occupies Morocco and Algeria and is the only member of its genus to reach Europe, where it occurs in the southern Iberian Peninsula. Genetic evidence indicates the Iberian populations were founded by colonizers from North Africa, and the two sides of the Mediterranean still show high levels of mitochondrial DNA admixture, particularly in the northern part of the range.

Decades of morphological study failed to bring order to this variation. Early taxonomists recognized two North African subspecies, while later authorities added a third from the Atlas Mountains and even suggested that one form deserved full species status. Others argued that all Moroccan populations represented a single, highly variable species whose differences simply reflected local adaptation. Molecular work then complicated matters further: mitochondrial DNA sequences revealed that the Tunisian species Acanthodactylus blanci and the form known as lineomaculatus both nest inside the erythrurus complex. One species-delimitation analysis of Algerian material suggested as many as 15 potential taxonomic units, the highest count recorded for any species examined in that study, alongside a striking east-west divergence across the Maghreb.

A major step toward resolution came in 2020, when researchers combining multi-locus sequence data with morphology recovered five major lineages and described two new species from the Atlas Mountains of Morocco: Acanthodactylus montanus from the High Atlas region and Acanthodactylus lacrymae from the Middle Atlas. These mountain forms are more closely related to Algerian and Tunisian lizards than to the bulk of Moroccan diversity, which remained parked in an unresolved cluster labeled the Ibero-Moroccan clade. Subsequent attempts to dissect that clade with mitochondrial and nine nuclear gene fragments managed only to distinguish four broad population groups separated by barriers to gene flow, while a recent genome-wide RADseq analysis of a contact zone hinted at strong reproductive isolation, suggesting even more cryptic species may be hiding within it.

To sharpen the geographic picture, the authors of the new study sequenced a partial fragment of the mitochondrial ND4 gene, together with adjacent transfer RNA regions, for 42 individuals, most of them from Morocco. The animals were caught in the field over a 20-year period, a small piece of tail tissue was removed for analysis, and each lizard was released at its collection site. Because single individuals often cannot be identified to lineage by appearance alone, every animal was sequenced. The new sequences were combined with 105 previously published ones from public databases, producing an aligned dataset of 769 base pairs, and four other Acanthodactylus species served as outgroups to root the analysis.

The laboratory workflow followed established protocols. Total genomic DNA was extracted from alcohol-preserved tail tissue using standard high-salt methods, and the ND4 fragment was amplified with the polymerase chain reaction under a regime of 33 cycles with an annealing temperature of 47 degrees Celsius. Sequences were edited and aligned with ClustalW, uncorrected genetic distances were calculated, and the team reconstructed evolutionary relationships using two complementary methods: Maximum Likelihood with 5,000 bootstrap replicates and Bayesian Inference run for five million generations in two independent analyses. The two approaches produced nearly identical trees, differing only slightly at the deepest nodes, and both recovered the same five major clades identified in earlier work.

The surprises came from geography rather than topology. New samples from the Debdou region, in the Moulouya river valley, filled a wide gap in the known distribution of the Ibero-Moroccan clade, which had previously been confirmed near Taza and then only by a single record more than 200 kilometers away in eastern Algeria. This matters because the Moulouya valley is widely regarded as a major biogeographic barrier that splits North African herpetofauna into western and eastern forms, and the lizards east of the valley do form a distinct subgroup within the clade, underscoring just how intricate the phylogeography of this complex really is.

Even more striking were the findings for the two Atlas Mountain species. Samples of Acanthodactylus lacrymae from a locality just north of Aït Aïssa extend that species’ range more than 100 kilometers to the east of previously confirmed populations. Meanwhile, a single specimen assigned to Acanthodactylus montanus pushes that lineage roughly 200 kilometers to the northeast and, crucially, places it within about 10 kilometers of the known range of A. lacrymae. The two species, previously thought to be broadly separated from each other, can therefore no longer be considered highly allopatric. That specimen also proved genetically remarkable: although phylogenetic analysis strongly supports it as the sister lineage to A. montanus, it differs from southern A. montanus samples by around 8 percent at the ND4 marker, a divergence exceeding that recorded between some accepted lacertid lizard species, though comparable to divergence levels seen within another Moroccan lacertid that showed incomplete lineage sorting in its nuclear genes.

The authors caution that mitochondrial DNA alone cannot settle these questions. Because the two Atlas species are extremely similar in appearance and cannot be reliably separated in the field, extensive genetic sampling across their contact zone will be needed to determine whether the two lineages occur in strict sympatry or whether there is genetic admixture between them. Genome-wide RADseq approaches, which have already revealed cryptic species within the Ibero-Moroccan clade, will likely be required to establish the taxonomic status of the newly detected montanus-like lineage as well. The study’s broader message is that both more genetic markers and more geographic sampling are essential, particularly across the southeastern part of Morocco, where the ranges of the different species and lineages remain incompletely mapped. For a group that has puzzled scientists since the nineteenth century, the fringe-toed lizards of North Africa are only slowly giving up their secrets, and each new map drawn from their DNA reveals just how much biodiversity still awaits formal recognition in the mountains and valleys of the Maghreb.

Subject of Research: Mitochondrial DNA diversity and geographic distribution of lineages within the Acanthodactylus erythrurus lizard species complex in North Africa

Article Title: New data on the diversity and distribution of lineages of the Acanthodactylus erythrurus species complex in North Africa derived from mitochondrial DNA markers

Article References: Harris, D. J., Sergiadou, D., & Faria, J. F. (2024). New data on the diversity and distribution of lineages of the Acanthodactylus erythrurus species complex in North Africa derived from mitochondrial DNA markers. Herpetozoa, 37, 281-286. https://doi.org/10.3897/herpetozoa.37.e118175

Image Credits: AI Generated

DOI: 10.3897/herpetozoa.37.e118175

Keywords: Acanthodactylus erythrurus, fringe-toed lizards, mitochondrial DNA, phylogeography, Morocco, North Africa, Atlas Mountains, cryptic species, Lacertidae, ND4 gene, species delimitation, herpetology

The study, published in Theoretical and Applied Climatology, was led by Nacer Aderdour of Hassan II University of Casablanca, together with colleagues in Casablanca and at the University of Bern. Its target was the six-month Standardized Precipitation Index, or SPI-6, one of the most widely used yardsticks of meteorological drought. The index condenses accumulated precipitation anomalies into a single number, allowing scientists and water managers to say, in a standardized way, whether a region is sliding into abnormally dry conditions. Forecasting that index ahead of time — one, two, or three months into the future — is the essence of drought early warning, and it is exactly where the new research draws its sharp conclusions.

The centerpiece of the work is a Convolutional Gated Recurrent Unit, or ConvGRU, a neural network designed to learn patterns that unfold both across space and through time. The architecture descends from the Convolutional LSTM introduced a decade ago for precipitation nowcasting, and it treats a sequence of gridded environmental observations the way a video model treats frames of film: each time step is a stack of images, and the network learns how features in one pixel neighborhood evolve into the next. The gated recurrent unit is a leaner relative of the LSTM, with fewer internal parameters, which makes it attractive when training data are limited — as they almost always are in drought research, where reliable records span only a few decades.

Feeding the ConvGRU was a rich stack of 18 input channels assembled from some of the most trusted observational products in climate science. Satellite-based CHIRPS precipitation estimates, which blend infrared imagery with ground station data, provided the rainfall backbone. ERA5-Land, a high-resolution reanalysis produced by the European Centre for Medium-Range Weather Forecasts, contributed atmospheric and land-surface fields. Satellite-derived NDVI, the Normalized Difference Vegetation Index from MODIS, tracked the green-up and browning of vegetation across the landscape. Static terrain features captured the formidable topography of the Atlas ranges. All of this was rendered onto a grid of 115 by 210 pixels at 5.5-kilometer resolution, covering the High Atlas and Anti-Atlas, with a strict leakage-free temporal split ensuring the models were tested only on periods they had never seen.

Against this deep learning heavyweight, the researchers lined up a set of challengers from a very different machine learning tradition: tree-based ensembles, most notably XGBoost, a gradient-boosted method that makes no pretense of understanding spatial structure. XGBoost builds a forest of decision trees sequentially, each one trained to correct the errors of the last, and it has become something of a workhorse in environmental prediction precisely because it is fast, robust, and hard to beat on tabular data. The benchmark also included damped persistence — the deceptively tough baseline that simply assumes next month’s conditions will resemble this month’s, slightly faded — and a ridge regression model.

At a one-month lead time, the verdict was a statistical dead heat. The ConvGRU achieved a root mean square error of 0.696, while the non-spatial XGBoost ensemble scored 0.698 — a difference so small it amounts to a rounding artifact. Both clearly beat the alternatives: the deep learning model improved on damped persistence by 11.8 percent, a margin the authors confirmed as statistically significant using a Diebold-Mariano test with a p-value of 0.003, and it outperformed ridge regression by 9.7 percent, the simpler linear model posting an RMSE of 0.771. But the headline finding is what did not happen: the added architectural complexity of spatiotemporal deep learning did not translate into superior forecast skill in this data regime. Across all forecast horizons, the gradient-boosted ensemble matched the convolutional models.

The real-world stakes of that one-month skill became vivid in a test-period evaluation of the severe drought of April 2024, an event that affected 92 percent of the study area. Asked to anticipate that drought one month ahead, the ConvGRU recovered 81 percent of the drought’s eventual extent, while XGBoost recovered an impressive 94 percent. Damped persistence, by contrast, managed only 29 percent — a striking demonstration that the machine learning models were capturing genuine predictive signal rather than merely extrapolating the recent past. For water managers deciding whether to trigger rationing, support farmers, or release reservoir water, that difference between 29 and 94 percent is the difference between reacting to a crisis and preparing for one.

Yet the study’s most consequential finding lies in what happens when the forecast horizon stretches further. At two and three months ahead, the predictive skill of every architecture — deep learning and tree-based alike — decays rapidly toward the damped persistence baseline. The mountains of the Atlas, it turns out, impose a hard practical limit on meteorological drought predictability: roughly one month. No amount of network sophistication, at least with the data currently available, extends that window. This is a humbling result in an era when deep learning has produced eye-catching successes elsewhere in the geosciences, from multi-year ENSO forecasts to skillful precipitation nowcasting, and it underscores a recurring lesson in machine learning applications: more parameters and fancier inductive biases do not automatically conjure information that the atmosphere and the observational record simply do not contain.

Why might the non-spatial model keep pace with its convolutional rival? The authors’ data regime offers a plausible explanation. Drought at the six-month accumulation scale is a slow, heavily smoothed quantity; much of its month-to-month evolution is governed by persistence and by large-scale anomalies that a per-pixel model can exploit without explicitly modeling spatial neighborhoods. With limited training samples — a handful of decades of monthly fields — a ConvGRU’s capacity to learn subtle spatial dynamics may go unrealized, while its extra complexity raises the burden of training and regularization. In such settings, gradient-boosted trees, which excel at squeezing signal from modest datasets, can match or exceed deep networks. The result does not diminish spatiotemporal deep learning; it delineates the conditions under which its advantages materialize.

For Morocco, where drought has repeatedly stressed agriculture, oases, and the mountain catchments that feed major aquifers and river basins, the operational implications are concrete. A reliable one-month early warning window is achievable today, and the tools to deliver it need not be exotic: a well-tuned tree-based ensemble fed with satellite precipitation, reanalysis fields, and vegetation data performs on par with the most elaborate architecture tested. The study’s code, pre-trained ConvGRU weights, and reproduction notebook have been released openly on Zenodo and GitHub, lowering the barrier for other drought-prone regions to replicate the benchmark. The broader message for the fast-growing field of AI-driven climate services is equally clear: before investing in ever-deeper spatiotemporal networks, measure honestly against simple baselines, respect the physics of predictability limits, and let the data — not the architecture — decide how far into the future a warning can honestly reach.

Subject of Research: Machine learning forecasting of meteorological drought in Morocco's Atlas mountains

Article Title: Early warning of meteorological drought in Morocco’s Atlas mountains: a multi-paradigm benchmark of satellite-augmented deep learning and tree-based models

Article References: Aderdour, N., Maanan, M., Rueff, H., & Rhinane, H. (2026). Early warning of meteorological drought in Morocco’s Atlas mountains: a multi-paradigm benchmark of satellite-augmented deep learning and tree-based models. Theoretical and Applied Climatology, 157(11), Article 700. https://doi.org/10.1007/s00704-026-06633-w

Image Credits: AI Generated

DOI: 10.1007/s00704-026-06633-w

Keywords: drought forecasting, Morocco, Atlas Mountains, deep learning, ConvGRU, XGBoost, Standardized Precipitation Index, CHIRPS, ERA5-Land, NDVI, early warning systems, machine learning

The team used inter-simple sequence repeat markers, known as ISSR markers, to probe the genomes of the collected trees. ISSR markers target short repeated DNA sequences scattered throughout the genome, amplifying the stretches between them to produce a fingerprint of bands that can be compared across individuals. Because these microsatellite-flanking regions mutate relatively quickly, ISSR markers are sensitive tools for detecting variation even among closely related cultivars of the same crop. In this study, 14 ISSR primers generated 214 scorable bands across the 93 cultivars, and a striking 184 of those bands, or 85.98 percent, were polymorphic, meaning they differed among at least some of the trees sampled.

Those numbers translate into exceptionally high genetic diversity for a cultivated fruit tree. The researchers calculated a resolving power of 8.00 for their marker set, a Nei’s gene diversity index of 0.34, a polymorphism information content of 0.52, and a Shannon information index of 0.34. Each of these metrics describes a different facet of variation, from how well the markers distinguish individual genotypes to how much uncertainty remains when guessing the genetic makeup of a randomly chosen tree. A PIC value above 0.5 is generally considered highly informative, indicating that the ISSR system captured substantial discriminating power across the Moroccan fig collection.

Perhaps the most consequential finding came from the analysis of molecular variance, or AMOVA, a statistical framework that partitions genetic variation into components within and between groups. The results showed that 89 percent of the total genetic variation resided within groups of cultivars, while only 11 percent was found between groups. In practical terms, this means that a farmer’s orchard in one valley may contain nearly as much hidden genetic variation as an entire region hundreds of kilometers away. Traditional agroecosystems, where multiple varieties are grown side by side and exchanged informally among neighbors, appear to be incubators of diversity rather than mere repositories of it.

Cluster analysis reinforced this picture of diffuse, locally rich diversity. Using UPGMA, a hierarchical clustering method that groups genotypes according to overall band-sharing similarity, the researchers organized the 93 cultivars into five main groups. The edible female figs from the Beni Mellal-Khenifra region showed particularly high genetic dispersion, scattering across the clustering diagram rather than gathering into a tight regional cluster. That dispersion suggests extensive exchange of planting material, seedling selection by farmers, and possibly multiple historical introductions of fig germplasm into the region, which historically served as a crossroads for trans-Saharan trade routes.

The caprifigs told a very different story. Caprifigs are the male, inedible figs that produce pollen and host the tiny fig wasp, Blastophaga psenes, which pollinates female fig flowers in a symbiosis so ancient and so intimate that neither partner can reproduce without the other. In the study, caprifigs formed a distinct and homogeneous cluster regardless of where they had been collected, separating cleanly from the edible female cultivars. This clear genetic split mirrors the gynodioecious breeding system of Ficus carica, in which male and female functional forms are genetically differentiated, and it validates the use of molecular markers to distinguish caprifig germplasm for breeding and pollination management.

The significance of this work extends well beyond Morocco’s borders. The fig is one of the oldest domesticated fruit trees in the world, and its center of diversity lies in the Middle East and Mediterranean basin, with Morocco recognized as a rich secondary reservoir. Earlier studies using microsatellite and other marker systems have documented diversity in Turkish, Tunisian, Algerian, and Jordanian fig germplasm, and Moroccan collections have been characterized morphologically and with SSR markers in previous work. What this new study adds is a broad, marker-based survey that explicitly spans both the mountain agroecosystems of the Atlas and the desert oases of the south, two environments that impose very different pressures on the trees growing in them.

That environmental breadth matters for conservation planning. Oases are fragile systems threatened by water scarcity, salinization, and rural depopulation, while mountain terraces face their own pressures from land abandonment and climate shifts. When genetic variation is distributed mostly within rather than between sites, losing any single site erodes a meaningful slice of the species’ total diversity. The authors argue that their results underline the value of ISSR markers for conservation and breeding programs, providing a cost-effective way to identify unique genotypes, avoid duplicating accessions in germplasm collections, and select parents for crosses that maximize genetic gain in fruit quality, drought tolerance, or disease resistance.

The study also fits into a growing body of work by the same research group at Sultan Moulay Slimane University, which has applied ISSR markers to Moroccan walnuts, plums, apples, and the endemic resin spurge Euphorbia resinifera. Together with companion studies by Mardoume and colleagues on the sugar profiles, organic acids, antioxidant activity, and phenotypic variability of Moroccan figs, the molecular data sketch a crop that is not only genetically rich but also chemically and morphologically diverse. For breeders seeking climate-resilient cultivars, such multidimensional diversity is raw material that cannot be recreated once lost.

For now, the message from the oases and mountains of Morocco is a hopeful one: centuries of traditional farming have quietly assembled one of the most genetically diverse fig populations ever surveyed with molecular markers. The challenge ahead is to document and conserve that diversity before the agroecosystems that sustain it disappear. With 85.98 percent of their marker bands polymorphic and nearly nine-tenths of the variation locked within local groups, Moroccan figs demonstrate that the future of food security under global change may depend as much on the smallholdings of traditional farmers as on any high-tech gene bank.

Subject of Research: Genetic diversity of fig tree (Ficus carica) cultivars and caprifigs in Moroccan oases and Atlas regions assessed with ISSR molecular markers

Article Title: Genetic diversity of Caprifigs and female fig (Ficus carica) from Moroccan oases and Atlas Regions revealed by ISSR markers

Article References: Mardoume, I., Bouda, S., Khachtib, Y., Zaggoumi, H., Neg, I., Ait Bella, Y., & Haddioui, A. (2026). Genetic diversity of Caprifigs and female fig (Ficus carica) from Moroccan oases and Atlas Regions revealed by ISSR markers. Plant Biosystems, 160(4), Article 216. https://doi.org/10.1007/s44473-026-00221-6

Image Credits: AI Generated

DOI: 10.1007/s44473-026-00221-6

Keywords: Ficus carica, genetic diversity, ISSR markers, Morocco, caprifigs, oases, Atlas Mountains, AMOVA, agroecosystems, germplasm conservation, plant breeding, molecular markers